基于局部积加权对比的红外弱小目标检测

针对复杂背景下因像素点噪声及高亮边缘干扰导致的对红外弱小目标检测率低、虚警率高的问题,提出一种基于局部 积加权对比的红外弱小目标检测算法。 首先,分别计算目标区域与背景区域均值,并得到目标与局部背景的差异性;提出一种 局部积加权方法,极大增强了小目标的显著性与抑制背景杂波的能力;其次,采用多尺度算法增强算法的自适应能力;最后,对 显著性图像进行自适应阈值分割,得到待检测的真实目标。 仿真实验结果表明,所提算法的信杂比增益( SCRg)和背景抑制因 子(BSF)相比现有算法均有一定提升,在复杂背景及强噪声干扰下仍具有良好的准确性和鲁棒性,实现了提高检测率,降低虚 警率的目的。

Infrared dim small target detection based on local product weighted contrast

An infrared dim small target detection algorithm based on local product weighted contrast is proposed for the low detection rate and high false alarm rate of infrared dim small targets in complex backgrounds caused by pixel noise and high-bright edge interference. First, the mean value of the target area and the background area is calculated respectively, and the difference between target and local background is obtained. A local product weighting method is proposed, which greatly improves the salience of small targets and the suppression ability of background clutter. Second, multi-scale algorithm is used to enhance the adaptive ability of the algorithm. Finally, adaptive threshold segmentation is performed on the saliency image to obtain the real target to be detected. Simulation results show that compared with the existing algorithms, SCRg and BSF of the proposed algorithm are improved to a certain extent, and still have good accuracy and robustness under the complex background and strong noise interference, achieving the purpose of improving the detection rate and reducing the false alarm rate.